A LoRA trained on Ume Aoki's art, best known as the character designer behind Madoka Magica, using the NAI model. No trigger words—just plug and play.
V2.0 should make it a bit easier to summon characters from Madoka Magica and Hidamari Sketch; just type the name and some relevant tags. See the images for examples.
Prefer textual inversion?
https://civarchive.com/models/68234/ume-aoki-textual-inversion
You can also find me on pixai.